Enantioselective synthesis of an atropisomeric diaryl ether

Article abstract of DOI:10.1002/anie.200705660

(Chemical Equation Presented) Twisted ethers: Introduction of a bulky alkylsulfinyl substituent ortho to the C-O axis of a diaryl ether imposes a powerful conformational preference (see scheme). The preference persists upon oxidation of the sulfoxide to a sulfone, leading to dynamic thermodynamic resolution of the atropisomeric ether. This is the first enantioselective synthesis of an atropisomeric diaryl ether not forming part of a macrocyclic ring.

Full text of DOI:10.1002/anie.200705660

Communications
DOI: 10.1002/anie.200705660
Atropisomeric Ethers
Enantioselective Synthesis of an Atropisomeric Diaryl Ether**
Jonathan Clayden,* Christopher P. Worrall, Wesley J. Moran, and Madeleine Helliwell
Members of the diaryl ether family of compounds have
diverse functions as mammalian hormones (thyroxine),[1]
powerful antibiotics (vancomycin, teicoplanin, ristocetin),[2]
and ligands for transition-metal-promoted hydroformylation
[3]
(bis(2-diphenylphosphinophenyl)ether;DPEphos).
They
are present in a range of natural products including the
bastadins, perottetines, riccardin B, and cyclic peptide K3.[4]
Stereochemical interest in diaryl ethers is focused on their
perpendicular conformation[1,5] and on their ability to exhibit
atropisomerism, particularly when constrained within a
macrocyclic ring, as they are in the vancomycin antibiotics.[2,6]
Recently we showed that atropisomerism is a general feature
of diaryl ethers 1 even outside of macrocyclic rings;[7] ethers 1
are generally resolvable, chiral compounds provided that
a) they bear at least three substituents ortho to the ether axis
and b) two of those substituents are bulky (tert-butyl group or
similar). Atropisomeric compounds have proved to be
supremely versatile, particularly as chiral ligands for metals,
yet chiral biaryl ethers remain an unexplored and under-
exploited class of atropisomers, not least because—in stark
contrast with the biaryls—no method yet exists for their
asymmetric synthesis.
Scheme 1. A strategy for the asymmetric synthesis of atropisomeric
diaryl ethers under thermodynamic control. a)Incorporate auxiliary to
govern bond orientation. b)Transform auxiliary into bulky R 2 group
while avoiding rotation.
required ethers by the route shown in Scheme 2. Nucleophilic
aromatic substitution of 3-bromo-2-chlorobenzonitrile (3)[7]
by 2-tert-butylphenoxide gave ether 4, which was converted
into sulfoxides by using either diacetonylglucose (DAG)
sulfinate esters 12 or isopropyl isopropylthiosulfinate (iPrS-
S(O)iPr).[16] The cyano group of 4 withstands bromine–
lithium exchange at low temperature, giving sulfoxides 5a
and (SS)-5b. Alternatively, the cyano group was converted by
a three-step sequence into the methoxymethyl group of 7,
which also underwent clean bromine–lithium exchange and
sulfinylation with (SS)-12a, (RS)-12a, and (SS)-12b to yield
sulfoxides (SS)-8a, (RS)-8a, and (SS)-8b, respectively.
In this communication we report the first enantioselective
route to a diaryl ether. The strategy[8] we employ makes use of
ꢀ
thermodynamic control over the orientation of the Ar OAr’
bond (Scheme 1). Incorporation of a chiral substituent (in this
case, a sulfoxide) with the ability to govern the favored
orientation of a nearby bond—a “conformational auxil-
iary”—allows one conformer of intermediate 2 to predom-
inate. Subsequent transformation of the auxiliary into a bulky
substituent traps the major conformer as a single atropisomer
and transforms a conformational preference into an atropiso-
meric enantiomeric excess.[9,10] The overall transformation of
(ꢁ )-1 to (ꢀ)-1 can be termed a dynamic thermodynamic
resolution.[11]
Previous studies[15] suggested that alkylsulfinyl groups
with tertiary alkyl substituents were essential for high levels of
conformational control, but we were unable to make tert-
alkylsulfoxides directly from 4 or 7. We therefore treated 5
and 8 with a base to deprotonate a to the sulfur center, and
quenched the resulting anions with methyl iodide. a Alkyla-
tions of 8 were successful and yielded sulfoxides 9. With 5,
however, the electron-withdrawing cyano substituent led to
ortholithiation of the sulfoxide;[17] methylation and subse-
quent in situ lateral lithiation of the sulfoxide yielded the
ethyl-substituted ethers 6.
The versatility of sulfoxide chemistry,[12] along with the
[13]
ꢀ
powerful dipole associated with the S O bond, prompted
us to choose a sulfinyl group as a means of achieving
conformational control. Sulfinyl groups have previously
performed well in controlling functional group orientation
in aromatic amides[10,13,14] and ureas.[15] We chose to make the
[*] Prof J. Clayden, C. P. Worrall, W. J. Moran, Dr. M. Helliwell
School of Chemistry
ꢀ
Diastereoisomeric conformers about the Ar OAr’ bond
University of Manchester
Oxford Rd., Manchester M13 9PL (UK)
Fax: (+44)161-275-4939
of ethers 2 interconvert slowly enough on the NMR timescale
1
for their conformational ratio to be detectable by H NMR
spectroscopy at ambient temperature,[7,18] but the selectivity
for one conformer over the other in 5, 6, and 8 was
disappointingly low (Table 1, entries 1–8). Nonetheless,
taking the lead from a clear trend towards greater selectivity
with more hindered substituents, we ortholithiated and
E-mail: clayden@man.ac.uk
[**] We are grateful to the EPSRC and GlaxoSmithKline for support, and
to Dr. Simon Peace for helpful discussions.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
3234
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3234 –3237

Angewandte
Chemie
gave crystals suitable for X-ray crystallography
analysis,[19] and for reasons described below we
ꢀ ꢀ
presume that their major Ar O Ar’ conformer in
solution corresponds to that in the solid state. The
X-ray crystal structures of (SS)-10a and (SS)-10b
ꢀ ꢀ
are shown in Figure 1, and the presumed Ar O
Ar’ conformation of 10 and 11 is indicated in
Figure 2.
We propose that a combination of steric and
electronic effects underlie the conformational
preference of 10 and 11. Dipole repulsion between
ꢀ
ꢀ
the S O bond and the proximal Ar O bond of the
ether, in combination with steric repulsion
between the aryl and alkyl substituents of the
sulfoxide group, favors conformations in which a
lone pair of electrons of the sulfoxide group
ꢀ
eclipses, or lies gauche to, the proximal Ar O
bond (Figure 2). Steric repulsion between the
ether substituents (the upper ring) and the alkyl
ꢀ
substituent on the sulfoxide group then favor Ar
ꢀ
O Ar’ conformer A over conformer B. Such a
rationale is in accord with the observed increased
selectivity afforded by bulkier substituents a and
ortho to the sulfoxide group.
Scheme 2. Synthesis of sulfinyl ethers. Reagents and conditions: a)Potassium
2-tert-butylphenoxide, DMF, 1508C, 16 h; b)1. nBuLi, THF, ꢀ788C, 1 min; 2. iPrS-
S(O)iPr, 12a, or 12b, ꢀ78!+208C, 16 h; c)1. LDA, ꢀ788C, THF, 0.5 h; 2. MeI,
ꢀ78–+208C, 3 h; d)DIBAL-H, tol, ꢀ78!+208C, 16 h; e)NaBH 4, THF, 30 min,
208C; f)NaH, MeI, THF, 20 8C, 30 min; g)1. nBuLi, THF, ꢀ788C, 1 h; 2. MeI,
ꢀ78!+208C, 2 h. DIBAL-H=diisobutylaluminum hydride; LDA=lithium
diisopropylamide.
We made some attempts to increase the level
of conformational control achievable in 2 by
synthesizing 13, 14, and 15 using routes analogous
to those used previously (Scheme 2). Despite their
Table 1: Conformational ratios in sulfinyl ethers 2.
Entry Struct.
R
R2’
R3 R6
Ratio[a] 2A:2B
1
2
3
4
5
6
7
8
5a
5b
6a
6b
8a
8b
9a
9b
10a
10b
11b
13a
13b
14a
14b
15
iPr
Cy
iPr
Cy
iPr
Cy
tBu
tBu
tBu
H
H
CN
CN
57:43
60:40
66:34
66:34
66:34
57:43
86:14
80:20
95:5
94:6
93:7
80:20
91:9
83:17
98:2
tBu Et CN
tBu Et CN
tBu
tBu
tBu
H
H
H
H
CH2OMe
CH2OMe
CH2OMe
CH2OMe
C(Me)(CH2)5 tBu
tBu
C(Me)(CH2)5 tBu Me CH2OMe
C(Me)(CH2)5 tBu Et CH2OMe
CMe2Et
CMe2Et
iPr
tBu
Cy
Me
Figure 1. a)X-ray crystal structure of ( SS)-10a; b)X-ray crystal structure
of (SS)-10b.
9
tBu Me CH2OMe
10
11
12
13
14
15
16
17
tBu
H
CH2OMe
tBu Me CH2OMe
iPr
H
H
H
H
CH2OMe
CH2OMe
CN
iPr
iPr[c]
iPr[c]
60:40
15
C(Me)2OMe 78:22
1
[a] Substituents refer to structure 2. [b] Measured by H NMR spectros-
copy in CDCl3. [c] With a Me group in the 6-position of the phenyl ring.
Cy =cyclohexyl.
Figure 2. Conformational preference in (SS)-sulfinyl ethers 10 and 11.
greater steric hindrance, 13a and 13b and 15 exhibited
conformational ratios inferior to the tert-butyl-substituted 9a,
10a and 5b. The same relationship between slightly decreased
steric hindrance and increased selectivity also held for 14,
with an iPr group at the 2’-position: 14b showed outstanding
conformational selectivity, possibly because of decreased
alkylated sulfoxides 9 to yield (SS)-10a, (RS)-10a, and (SS)-
10b. 10b was furthermore laterally lithiated and methylated
to yield ethyl-substituted (SS)-11b.
Remarkably, these alkylations increased the ratio of the
conformers to 15–20:1 in favor of 2 A. (SS)-10a and (SS)-10b
Angew. Chem. Int. Ed. 2008, 47, 3234 –3237
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3235

Communications
Table 2: Stereochemical data for ethers 17.
e.r.[a] [a]2D3[b] t = rac
DGrac
t = rac
[d]
1
1
Entry Ether
2
2
(408C)[c]
[kJmolꢀ1
]
(258C)[e]
[h]
[days]
1
2
3
P-(ꢀ)-
17a
M-(+)-
17a
90:10 ꢀ110.9 134
113.6
–
100
95:5 +106.2
–
–
P-(ꢀ)-
91:9 ꢀ172.9 260
111.8
50
17b
[a] Determined by HPLC methods with a chiral stationary phase
(Chiralcel OT(+)). [b] c=1.0, acetone. [c] Determined in 95:5 hexane/
CH2Cl2. [d] Determined at 40 8C. [e] Estimated half-life for racemization
at room temperature assuming DS° ꢂ0.
unfavorable interactions between the 2’ substituents and the
sulfinyl group (note the proximity of the two tert-butyl groups
in Figure 1). Combining this change with a more bulky
substituent at the 6 position or a 2’,6’-disubstituted distal ring
(necessary to maintain a rotational barrier sufficient to retain
atropisomerism) gave promising selectivity in the methylsulf-
oxide 16,[20] but at the cost of greatly reduced reactivity.
The final step in the sequence employing the sulfinyl
substituents as chiral auxiliaries must be the transformation of
the sulfoxide into an achiral substituent of sufficient steric
Figure 3. X-ray crystal structure of (P)-(ꢀ)-17b.
ꢀ ꢀ
bulk to prevent free rotation about the Ar O Ar’ axis and to
preserve any atropisomeric enantioselectivity achieved. We
elected to use a sulfonyl group to perform this function since
it could be introduced straightforwardly by oxidation of the
sulfoxide groups to sulfone groups.[21]
The preparation of ethers 17 represents the first enantio-
selective synthesis of an atropisomeric diaryl ether devoid of
stereogenic features other than the ether axis itself. It was
achieved by dynamic resolution under thermodynamic con-
trol, a strategy particularly suited to the synthesis of
atropisomers.
Accordingly, the three sulfoxides (SS)-10a, (RS)-10a, and
(SS)-10b were treated with mCPBA in CH2Cl2 at 08C to yield
atropisomeric ethers (P)-17a, (M)-17a, and (P)-17b
(Scheme 3). HPLC on a chiral stationary phase indicated
that the products were enantiomerically enriched (Table 2).
The X-ray crystal structure of (P)-17b[19] confirmed its
absolute stereochemistry and hence also our assignment of
the major sulfoxide conformer in solution (Figure 3). Atro-
pisomeric ethers 17 were stable to racemization at room
temperature, but showed a first-order decay in their enantio-
meric purity at 408C in 95:5 hexane/CH2Cl2 (half-lives shown
in Table 2). Extrapolation of these rates to 258C suggests
enantiomeric stability in solution over a timescale of weeks.
Received: December 11, 2007
Revised: January 4, 2008
Published online: March 17, 2008
Keywords: atropisomerism · conformation analysis ·
.
diaryl ethers · sulfur
[1] B. M. Duggan, D. J. Craik, J. Med. Chem. 1997, 40, 2259.
[2] K. C. Nicolaou, C. N. C. Boddy, S. Bräse, N. Winssinger, Angew.
Chem. 1999, 111, 2230; Angew. Chem. Int. Ed. 1999, 38, 2096.
[3] M. Kranenburg, Y. E. M. ven der Burgt, P. C. J. Kamer,
P. W. N. M. van Leeuwen, K. Goubitz, J. Fraanje, Organometal-
lics 1995, 14, 3081.
[4] E. A. Couladouros, E. N. Pitsinos, V. I. Moutsos, G. Sarakinos,
Chem. Eur. J. 2005, 11, 406;F. Theil, Angew. Chem. 1999, 111,
2493; Angew. Chem. Int. Ed. 1999, 38, 2345;J. S. Sawyer,
Tetrahedron 2000, 56, 5045.
[5] G. Montaudo, P. Finocchiaro, E. Trivellone, F. Bottino, P.
Maravigna, Tetrahedron 1971, 27, 2125;H. Kessler, A. Rieker,
W. Rundel, Chem. Commun. 1968, 475;J. J. Bergman, W. D.
Chandlwe, Can. J. Chem. 1972, 50, 353;P. A. Lehman, E. C.
Jorgensen, Tetrahedron 1965, 21, 363.
Scheme 3. Synthesis of atropisomeric ethers. Reagents and conditions:
a) mCPBA, CH2Cl2, 08C. The yield in the parentheses is based on
recovered starting material.
[6] D. L. Boger, J.-H. Weng, S. Miyazaki, J. J. McAtee, S. L. Castle,
S. H. Kim, Y. Mori, O. Rogel, H. Strittmatter, Q. Jin, J. Am.
3236
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3234 –3237

Angewandte
Chemie
Chem. Soc. 2000, 122, 10047;D. L. Boger, S. Miyazaki, O.
Loiseleur, R. T. Beresis, S. L. Castle, J. H. Win, Q. Jin, J. Am.
Chem. Soc. 1998, 120, 8920;D. J. Boger, S. Miyazaki, S. H. Kim,
J. H. Wu, O. Loiseleur, S. L. Castle, J. Am. Chem. Soc. 1999, 121,
3226;D. L. Boger, S. Miyazaki, S. H. Kim, J. H. Wu, S. L. Castle,
O. Loiseleur, Q. Jin, J. Am. Chem. Soc. 1999, 121, 10004;D. A.
Evans, M. R. Wood, B. W. Trottter, T. I. Richardson, J. C.
Barrow, J. K. Katz, Angew. Chem. 1998, 110, 2864; Angew.
Chem. Int. Ed. 1998, 37, 2700.
(SS)-sulfoxides. (RS)-12a was also available as a 96:4 diastereo-
isomeric mixture by the method of Fernandez et al., and was
used to make (RS)-sulfoxides. Ellmanꢀs tert-butyl tert-butylthio-
sulfinate (D. J. Weix, J. A. Ellman, Org. Lett. 2003, 5, 1317) is a
convenient precursor to tert-butylsulfoxides, but it failed to react
with organometallic derivatives of 4, preventing direct synthesis
of 9a and its derivatives.
[17] C. Quesnelle, T. Iihama, T. Aubert, H. Perrier, V. Snieckus,
Tetrahedron Lett. 1992, 33, 2625.
[7] M. S. Betson, J. Clayden, C. P. Worrall, S. Peace, Angew. Chem.
2006, 118, 5935; Angew. Chem. Int. Ed. 2006, 45, 5803.
[8] For a discussion of thermodynamic conformational control, see
J. Clayden, Chem. Commun. 2004, 127.
[9] For comparable thermodynamically controlled methods applied
to other classes of atropisomers, see reference [10] and J.
Clayden, P. Johnson, J. H. Pink, M. Helliwell, J. Org. Chem.
2000, 65, 7033;J. Clayden, L. W. Lai, M. Helliwell, Tetrahedron
2004, 60, 4399.
[18] In some cases the diastereoisomeric conformers exist as
separable atropisomers. For example, 8a showed two spots on
TLC;a second elution along a perpendicular axis showed cross-
spots indicating interconversion on the plate over a period of
minutes. Flash chromatography of 8a allowed us to obtain mixed
samples enriched in either diastereoisomer which both equili-
brated to the same mixture within 12 hours at room temperature.
All ratios reported in Table 1 were measured after ensuring
complete equilibration between diastereoisomers.
[10] J. Clayden, D. Mitjans, L. H. Youssef, J. Am. Chem. Soc. 2002,
124, 5266.
[11] P. Beak, D. R. Anderson, M. D. Curtis, J. M. Laumer, D. J.
Pippel, G. A. Weisenburger, Acc. Chem. Res. 2000, 33, 715.
[12] The Chemistry of Sulfur-containing functional groups (Eds.: S.
Patai, Z. Rapoport), Wiley-VCH, Weinheim, 1993.
[13] M. S. Betson, J. Clayden, M. Helliwell, D. Mitjans, Org. Biomol.
Chem. 2005, 3, 3898.
[14] J. Clayden, C. C. Stimson, M. Keenan, Synlett 2005, 1716.
[15] M. S. Betson, A. Bracegirdle, J. Clayden, M. Helliwell, A. Lund,
M. Pickworth, T. J. Snape, C. P. Worrall, Chem. Commun. 2007,
754.
[16] Diastereoisomerically pure DAG-sulfinate esters (SS)-12a and
(SS)-12b were made by the methods of Fernandez et al. (I.
Fernandez, N. Khiar, J. M. Llera, F. Alcudia, J. Org. Chem. 1992,
57, 6789) and of Alayrac et al. (C. Alayrac, S. Nowaczyk, M.
LemariØ, P. Metzner, Synthesis 1999, 669). Both react with
aryllithium sunstrates with inversion of configuration and a
change in priority at the stereogenic sulfur atom thus yielding
[19] CCDC 669805 ((SS)-10a), 669806 ((SS)-10b) and 669807 ((P)-
(ꢀ)-17b) contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre at www.ccdc.cam.ac.
uk/data_request/cif.
[20] Other methylsulfoxide have shown very low levels of conforma-
tional control (see reference [15]).
[21] In general, the success of the method relies on the fact that this
step of the sequence proceeds faster than the rate of intercon-
version of the conformers or that the rate of oxidation of the two
diastereoisomeric conformers are comparable. The observation
of slow diastereoisomeric interconversion in 8a suggests the
former holds true. Under these ideal conditions the e.r. of 17 is
the product of the e.r. of the starting sulfoxide 10 and its
conformational purity. For a discussion of the application of the
Winstein–Holness equation in comparable situations, see J. I.
Seeman, Chem. Rev. 1983, 83, 83;N. S. Zefirov, Tetrahedron
1977, 33, 2719.
Angew. Chem. Int. Ed. 2008, 47, 3234 –3237
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3237